Application of a digital filter for modelling river suspended sediment concentrations

Abstract

The relationship between suspended sediment concentration and discharge in a river system is the result of a complex interaction of factors within the basin and of specific storm events. Nevertheless, estimation of suspended sediment concentrations during storm events is essential for prediction of loadings from storm-controlled river basins to the receiving waters. The Venoge River, a tributary of Lake Geneva, Switzerland, with a basin area of 237 km 2, has been monitored during eight storm events between October 1986 and November 1987. A digital filter is used to simulate the suspended sediment response during individual events with the principal factors being antecedent flow, maximum discharge and time during which maximum flow was attained. The measured and calculated sediment concentration curves were assessed using a x 2-test. It was concluded that the two curves were similar in shape at 95% confidence interval. However, the model is not always capable of simulating the phase lead effect of the sediment distribution curve relative to the hydrograph. This problem is most evident during high-intensity storms at periods of low base flow; and consequently, loadings calculated for individual storms of this type could be overestimated. Fortunately, the model provides good predictions during less-intense, longer-duration events and therefore loadings calculated by summing events for an annual period will tend not to be overestimated. Further work is necessary to evaluate the model for additional events, in particular spring snow melt, and to determine its applicability to other river systems.